Japan

[Norselord]

I just recently moved up from Noble to Prince. Not too much different really. I pretty much always play random leaders, because I like exploring the different play styles. It also keeps me from falling into a rut, although i do have a tendency to go for cottage economy if i end up getting assigned financial, and end up going specialist if i get assigned philosphical. I also try to work around whatever my unique building or unique unit is, whether or not it is considered the "best" direction of play or not by the general community.

The last three games since moving up to Prince, I have been 'randomly' assigned to play as the Japanese. I have tended to avoid going for early wonders, trying only to snag the pyramids. After getting BW and AH, I Beeline courthouses via alphabet and currency, following that up with machinery and civil service before going right for gunpowder and steel.

The capital is used to pump out a settler and a few workers with support troops before building 'mids. Afterwards the cap becomes the science/GP city.

The two other types of cities I will run are military and science/money.

Military cities get granary, barracks, stables, forges, courthouses and any required health/happ buildings, and continuously pump out a stack of: defender, attacker, siege, attacker, spearguy, siege, attacker. and repeat. A slight variation on this city will also provide an occasional worker.

Wealth/science cities build monument, granary, library, marketplace, courthouse, grocer, and whatever health/happiness is needed. The build will either be wealth or science, depending on whether or not i need cash to upgrade units or tech to research towards new military units.

The ratio of Military cities to wealth/tech cities is normally 2:1.

The early economy and expansion is supported entirely through pillaging and capturing the first victim's country. Afterwards the specialist economy picks up the slack between wars.

I am experiencing problems in the late medieval/early renaissance era, and I can't figure out why. I consider myself adept at carrying on wars, using my mounted units & spies to deny critical war resources. Creating a mixed stack of units with effective use of siege weapons. I am usually capable of taking out 1-2 civs in the early game without a hitch. However by this time, there are usually 1 or 2 civs that have mobilized a larger armer or gained a significant tech lead.

I will typically run Rep/Bureau/Slave/Mercantilism/Theocracy for civics.

Am I playing the Japanese too aggresively? I figured the focus for these guys was to do early rushes, regroup buildup cannons/early gunpowder, take out a couple more civs, regroup build shale plants, build tanks/artillery/infantry, take out the last few civs.

The path to victory for the Japanese seems to clearly indicate a domination or conquest victory, which is not how i normally play. Is it a bad idea to regroup after every couple of civs are taken down, am i losing momentum while i try to rebuild the captured cities? Do i need to burn more cities (currently keep all of them)? Am i expanding too fast? Is my ratio of city types off?

Please provide advice, dear reader, on how to most effectively play with the Japanese. If you can, please include recommendations for: economy, city types, build orders, civic recommendations, conquering strategies, and any other pointers.

Thanks!

[Ming]

Burn more cities... Keep the few that are worth keeping... holy cities, wonders, key resources, well developed and good land.... First, you can roll through on offense faster without having to protect every city you take, and you don't totally trash your economy by keeping a bunch of "less than optimal" cities. AI's will build cities ANYWHERE. Many of them are pretty worthles... why keep them.

Also... you might want to consider caste system over slavery if you are going to build the mids and run rep. You can get specialists going without having to build the improvements. It can really jump start the beakers.

[wodan11]

Notes:
-- post screenshot of F2 screen!
-- all Med and High cost civics. Maybe consider HR if CE.
-- if SE then Bureau doesn't make sense, unless you cottage the capitol or turn it into a production city. And, if cottage it then can't use the culture slider for happy.
-- Mercantilism pbly should be switched to SP when get the tech especially if running Caste and SE. Also, consider creating some island vassals... you still get foreign trade routes under mercantilism and they can be big bucks plus remove your highest maint cities from your expenses.
-- military to wealth cities 2:1 makes me want to suggest CE under HR and Bureau, keep some friends and do free market.

[Norselord]

Burn more cities... Keep the few that are worth keeping... holy cities, wonders, key resources, well developed and good land.... First, you can roll through on offense faster without having to protect every city you take, and you don't totally trash your economy by keeping a bunch of "less than optimal" cities. AI's will build cities ANYWHERE. Many of them are pretty worthles... why keep them.

Also... you might want to consider caste system over slavery if you are going to build the mids and run rep. You can get specialists going without having to build the improvements. It can really jump start the beakers.

Seems to me that when I burn a city, some peaceful civ moves in and starts a city up on that spot, and this causes a discontinuity between my empire and the city's that ARE worth keeping (thanks for building all those wonders for me Charlie)

[Norselord]

Notes:
-- post screenshot of F2 screen!
-- all Med and High cost civics. Maybe consider HR if CE.
-- if SE then Bureau doesn't make sense, unless you cottage the capitol or turn it into a production city. And, if cottage it then can't use the culture slider for happy.
-- Mercantilism pbly should be switched to SP when get the tech especially if running Caste and SE. Also, consider creating some island vassals... you still get foreign trade routes under mercantilism and they can be big bucks plus remove your highest maint cities from your expenses.
-- military to wealth cities 2:1 makes me want to suggest CE under HR and Bureau, keep some friends and do free market.

There isn't a screenshot for F2, this is an amalgam of the last three games playing as Japan.

Since I am at war mostly non-stop i prefer to run SE instead of CE, since i don't want to loose revenue due to suicide incursions into my lands that result in pillaged towns, etc.

[Norselord]

Thanks for the responses, ofcourse most answers lead to more questions:
- synergistic civics to specialist economy at war? I like slavery for the increased production, caste system is fine for generating massive tech and GPs, but i run SE for being able to tech with a low slider, due to high cost of military maintenance.

- Is there anything wrong with running Police State/Nationhood/Slavery/State Propoerty/Theocracy?

- Do you agree that Japan should be played warmonger-style?

- If military:wealth/research should not be 2:1, what is the best ratio?

- should city assignments be divided up differently?

- Since Japan = Agg/Def, gunpowder units get three automatic promotions, it would seem that this has to be a key to the japanese playstyle.

- Any decent gambits for Japan?

- Is it viable to wait wait after an initial axerush to build up $$$ and troops until steel/gunpowder and only then start the big conquest, instead of continuously conquesting from the second the first 7 axemen are out?

It doesn't seem like my economy gets too trashed, it takes about 10-15 turns to research techs. Epic speed. In the early game the economy does tank a little, but i rely on revenue generated from sacking cities and pillaging to get me over the hump from currency to code of laws.

Are any of the cultural/religious techs worth researching at all? I would think that I should be able to demand these techs through superior military from the wussies that build culture/religion.

[wodan11]

There isn't a screenshot for F2, this is an amalgam of the last three games playing as Japan.

I understood that. My comment was shorthand (though it would be nice if you could post a couple of samples from various turns in the games) for "If you're asking for economic advice then you should be checking F2 frequently throughout your games... you'll get a feel for the black holes where all your commerce is being poured into. Once you know the culprit, you should be able to narrow down a solution. e.g., if city maint is killing you, then make sure you prioritize CoL for courthouses, maximize trade routes across your empire (and consider NOT running Mercantilism), etc."

Since I am at war mostly non-stop i prefer to run SE instead of CE, since i don't want to loose revenue due to suicide incursions into my lands that result in pillaged towns, etc.

It was obvious you're doing military heavy and nonstop wars, pretty much, from tips such as the 2:1 ratio of cities. Your oldest and most beneficial commerce cities will be at the heart of your empire, where they aren't subject to pillage. CE is easily a viable option for a military heavy game.

- synergistic civics to specialist economy at war? I like slavery for the increased production

Contrary to popular belief, Slavery does not really give increased production. It gives rushed production.

(Except in some few cases such as a water city with little else but lots of fish, but you're not going to be using those as your military unit generating cities in your Japan games.)

caste system is fine for generating massive tech and GPs, but i run SE for being able to tech with a low slider, due to high cost of military maintenance.

There's a false premise hidden there. Consider this: if the bottom line of CE and SE are fairly equivalent (which they are, or one would be broken and everybody would use that one), then the commerce equivalent (+) produced by each is the same. And, if a given game requires some # for unit maintenance, then that has to be subtracted from the total commerce equivalent. SE is able to do this on low slider, but CE on higher slider produces the same amount of gold. And the left over is the same in both cases.

What SE lets you do is run slider to get happiness which doesn't cost much or . CE can't do that.

However, SE does that at the cost of "fixing" the / sliders, effectively. They aren't really fixed, but you have to go through a lot of micromanagement to adjust scientists to merchants, for example. And, that's a pain. Plus, in some cases you might not have build Libraries or Markets (etc.) to save hammers. Which is another strength of SE but it's also a limitation.

- Is there anything wrong with running Police State/Nationhood/Slavery/State Propoerty/Theocracy?

Why in the world would there be something wrong with it? Creative civic choices are cool and show you know how to play the game in different ways.

Or do you mean are there any hidden costs or dangers? Well your civic costs are much lower than the previous choices. Police & Nationalism will be wonderful for the warmonger. I still think Slavery is dubious. SE is not so good a choice without Rep so I would suggest CE, in which case Emancipation is a good choice instead of Slavery. Prior to availability of Emancipation, perhaps Serfdom. Think of it this way: with Serfdom you can build half as many workers, which is a big help since unit maintenance is a problem for you already. And not building workers also boosts your cities because they can grow faster / earlier, and can make buildings earlier.

- Do you agree that Japan should be played warmonger-style?

Nope. I agree that it can be played warmonger style.

Two ways to look at Traits, UUs and UBs. One is to try to find synergies to leverage the benefit they provide. Two is to use them to "balance out" with weaknesses of strategy to make for a more well-rounded game.

That said, I find it really hard not to use Samurai to have a big mid-game war. But that doesn't mean I plan to be at war all game (= "warmonger"). But they can be used to get "breathing room" to x1.5 - x3 your empire size, with which you could do a very strong / early Space win, especially with the help of Shale Plants.

- If military:wealth/research should not be 2:1, what is the best ratio?

Clearly, you're getting into economic trouble, so either there's something you should be doing or not doing (such as razing more cities as rah suggested), or else this ratio should be backed off.

Personally, I don't think it's a good idea to have ironclad rules... I wouldn't want to suggest "if you warmonger than you should have a 2:1 city ratio". It depends from game to game. Terrain will affect it, as will a half dozen other things. Heck, terrain can force you to choose CE over SE or vice versa, which is a much bigger game changer than whether you have 8 miltary cities or 7. So it's not like all this can be planned.

But anyway, if you're currently doing 2:1 but having money trouble, then obviously one simple fix is to back off to 1.5:1 or so.

- should city assignments be divided up differently?

What do you mean... what buildings they make?

- Since Japan = Agg/Def, gunpowder units get three automatic promotions, it would seem that this has to be a key to the japanese playstyle.

Again, a false premise.

Let's look at an opposite example. Say... Darius / Persia. Should he (ORG/FIN) all but ignore military and focus on REX and cottage spamming? If he did, he's be an easy target for someone to conquer.

Same here... if as Toku / Japan we focus on warmongering, then we're at risk of crappy economy or falling way behind on tech.

At some level, you have to have a more balanced game. If you focus too much on one aspect, then you get in trouble.

- Any decent gambits for Japan?

A drill Samurai army is huge.

Also, getting Axes to 10XP from Barbs, then dropping to 0 slider as soon as you get Samurai to get money, and cash upgrade mass numbers of them at the same time you build more. (Or, you can ensure you get a Great Merchant or two and do trade missions for the gold.)

- Is it viable to wait wait after an initial axerush to build up $$$ and troops until steel/gunpowder and only then start the big conquest, instead of continuously conquesting from the second the first 7 axemen are out?

Well sure.

It doesn't seem like my economy gets too trashed, it takes about 10-15 turns to research techs. Epic speed. In the early game the economy does tank a little, but i rely on revenue generated from sacking cities and pillaging to get me over the hump from currency to code of laws.

You might also consider not wasting so much money early game on unit maintenance and instead build up infrastructure and tech to Currency and CoL faster.

Are any of the cultural/religious techs worth researching at all? I would think that I should be able to demand these techs through superior military from the wussies that build culture/religion.

Sometimes they won't trade or nobody has it when you need it. Again, each game is different... hard and fast rules are anything but.

[Thoth]

I just recently moved up from Noble to Prince. Not too much different really. I pretty much always play random leaders, because I like exploring the different play styles. It also keeps me from falling into a rut, although i do have a tendency to go for cottage economy if i end up getting assigned financial, and end up going specialist if i get assigned philosphical. I also try to work around whatever my unique building or unique unit is, whether or not it is considered the "best" direction of play or not by the general community.

The last three games since moving up to Prince, I have been 'randomly' assigned to play as the Japanese. I have tended to avoid going for early wonders, trying only to snag the pyramids. After getting BW and AH, I Beeline courthouses via alphabet and currency, following that up with machinery and civil service before going right for gunpowder and steel.

The capital is used to pump out a settler and a few workers with support troops before building 'mids. Afterwards the cap becomes the science/GP city.

The two other types of cities I will run are military and science/money.

Military cities get granary, barracks, stables, forges, courthouses and any required health/happ buildings, and continuously pump out a stack of: defender, attacker, siege, attacker, spearguy, siege, attacker. and repeat. A slight variation on this city will also provide an occasional worker.

Wealth/science cities build monument, granary, library, marketplace, courthouse, grocer, and whatever health/happiness is needed. The build will either be wealth or science, depending on whether or not i need cash to upgrade units or tech to research towards new military units.

The ratio of Military cities to wealth/tech cities is normally 2:1.

The early economy and expansion is supported entirely through pillaging and capturing the first victim's country. Afterwards the specialist economy picks up the slack between wars.

I am experiencing problems in the late medieval/early renaissance era, and I can't figure out why. I consider myself adept at carrying on wars, using my mounted units & spies to deny critical war resources. Creating a mixed stack of units with effective use of siege weapons. I am usually capable of taking out 1-2 civs in the early game without a hitch. However by this time, there are usually 1 or 2 civs that have mobilized a larger armer or gained a significant tech lead.

I will typically run Rep/Bureau/Slave/Mercantilism/Theocracy for civics.

Am I playing the Japanese too aggresively? I figured the focus for these guys was to do early rushes, regroup buildup cannons/early gunpowder, take out a couple more civs, regroup build shale plants, build tanks/artillery/infantry, take out the last few civs.

The path to victory for the Japanese seems to clearly indicate a domination or conquest victory, which is not how i normally play. Is it a bad idea to regroup after every couple of civs are taken down, am i losing momentum while i try to rebuild the captured cities? Do i need to burn more cities (currently keep all of them)? Am i expanding too fast? Is my ratio of city types off?

Please provide advice, dear reader, on how to most effectively play with the Japanese. If you can, please include recommendations for: economy, city types, build orders, civic recommendations, conquering strategies, and any other pointers.

Thanks!

First off, you'll get better (ie specific) advice if you post a save for people to look at. General advice in Civ4 is always a bit iffy, as every game is different based on the land you have to work with and the leaders in play.

Toku is a tough leader. Agg and Pro are both quite weak traits that provide little/no help with the economy from traits, his only cheap buildings are barracks, walls and castles.

Early techs: Toku starts with Agriculture and the Wheel the two most expensive first tier techs (edit: he starts with FISHING and the wheel. :doh: ) and they are both worker techs, which helps give him a leg up in the very early game. Your worker will never lack for something to do. Fishing is situational, if you are coastal with seafood specials, it's very nice to start with, if you're inland it's a lot less useful (other than being an optional prerequisite for Pottery)

Teching mining/bronze as your first two techs is often not the strongest opening. If you have cows/pigs or sheep in your starting city's fat cross, teching Animal Husbandry as your first tech and improving these resources will get you off to a much stronger start than teching to Bronze. If you have a coastal capital with multiple seafood resources, fishing first is often your best opening.

In virtually all cases, making your first build a worker will get you off to a stronger start than any other first build will.

Civics/city development:

If your capital is suitable for development as a GP farm, consider moving your palace to your best commerce (ie heavily cottaged) city. B-cracy gives you a 50% boost to hammers and a 50% boost to your base commerce. If your capital is running a lot of specialists to generate GPs, you get very little bonus from B-cracy. OTOH, a capital with 12-14 cottages (work 'em as early as practical) gets a huge boost from B-cracy, especially once you get an academy in. Try farming your great scientists in one of your other cities so your capital can work it's cottages.


If you are running a specialist heavy game, consider running caste instead of slavery. Caste lets you turn any city with 2+ food resources into a mini-GP farm.

A "typical" (ha, ha) empire set up for me has 2 commerce cities (capital plus one other city with rivered grass/floodplains) that get heavy cottaging, 1 GP farm (look for a site with 2-3 food resources and ideally a few hills and some forests for building/chopping the Great Library/National Epic), and the rest of the cities being production oriented (either via mines or the whip). I may also run a secondary GP farm early to help generate 1-2 extra Great Scientists (for lighbulbing) or Great Merchants (for trade missions to fund deficit research/military upgrades). Once the main GP farm is up and humming, I'll convert the secondary farm to a Globe powered whipping/drafting city.

Early courthouses are vastly over-rated. I generally don't build courts until a city is costing ~8 gpt (after accounting for inflation). Courts are very expensive (120h at normal speed), if it is only going to save you 2-3 gpt you are better off building wealth/research with those hammers instead.

In terms of military composition (outside of early rushes) aim for about 50% seige/30% best attacker/20% stack defence. So a typical late classical/early medieval stack (for me) would consist of ~12 cats/6-7 swords/2-3 axes/a few spears if the target has access to horses/plus 1-2 chariots or HAs. Having a few mounted units with your stack gives you the chance to pick off stray workers or isolated units without having to divert your main stack from it's objective.


Suggestion: Start up a game as Toku using your usual settings and post the 4000BC save here for people to discuss and/or play shadow games.

[Krill]

Only build wonders that fit the strategy you are employing, and tailor your strategy to fit the map.

[Norselord]

Early techs: Toku starts with Agriculture and the Wheel the two most expensive first tier techs (edit: he starts with FISHING and the wheel. :doh: ) and they are both worker techs, which helps give him a leg up in the very early game. Your worker will never lack for something to do. Fishing is situational, if you are coastal with seafood specials, it's very nice to start with, if you're inland it's a lot less useful (other than being an optional prerequisite for Pottery)

When I read that originally, I was confused, because I remember in one of my games that I had two seafood resources and that my first build was a workboat. I assumed that I was wrong, and might have gotten the tech from a goodie hut or something right off the bat.

When on water, the Main City build order:
1) Workboat
2) Warrior
3) Worker
4) Warrior (unless one popped from a hut, or my 1st warrior is finished exploring)
5) Settler
6) warrior
7) Settler
8) Pyramids

Research order is:
1) Mining
2) Bronzeworking
3) Masonry
4) Agriculture
5) Animal Husbandry
6) Hunting
7) Archery
8a) Horsebackriding if horses are available
8b) Pottery
9) Writing
10) Alphabet
11) Currency
12) Code of Laws
13) Iron Working
14) Metalcasting
15) Machinery
16) Civil Service
17) Construction
18) Guilds
19) Engineering
20) Paper
21) Education
22) Gunpowder
23) Chemistry
24) Steel

Try to demand/espionage/trade non-wonder, non-military unit techs for: Mysticism, Meditation, Polytheism, Priesthood. Monotheism, Aesthetics, Literature, Calendar.

Bear in mind that the above research tree only applies IF i am playing as Toku, and is subject to change pending conditions. It is only an objective guide that I have used to head the ship of state towards domination, not a rigid set of rules. However, for the sake of improving my game, please let me know an improved order for research, and let me know if I am skipping a critical tech. If you can provide the reasons as well, then I will not only have a fish, but will also be able to fish.

Thanks again for the advice!

[Thoth]

Contrary to popular belief, Slavery does not really give increased production. It gives rushed production

Close, but not quite.

Slavery allows you to convert surplus food into hammers. In a high food surplus city it will increase production significantly. (assuming of course, that the whips are for 2+ pop and that the tiles being whipped away are not high production tiles).

I still think Slavery is dubious

Sometimes. Kremlin powered whipping from biology farms can produce mad production, but I'm inclined to agree that SP/caste workshops generally perform better.

Otherwise, excellent advice as usual.

[Thoth]

When I read that originally, I was confused, because I remember in one of my games that I had two seafood resources and that my first build was a workboat. I assumed that I was wrong, and might have gotten the tech from a goodie hut or something right off the bat.

When on water, the Main City build order:
*snip*

Yeah, sorry about the bad info, it had been awhile since I've played Toku.

Initial build order really depends on the resources available. With a couple of land based food resources, it is often better to build a worker 1st while researching AG/AH. A workboat on Epic costs 45 hammers and can improve 1 tile. A worker costs 90 food/hammers but can improve many tiles.

Which one is better really depends on the actual in-game situation. If I've got wheat/corn in my fat cross and a non-forested hill then I'll almost always go worker-workboat (assuming coastal with seafood).

Unless you have stone, consider skipping the Pyramids. It's a very expensive wonder to build early, and often you are better off investing those hammers in more settlers/workers or an army to stomp a neighbour. Heriditary rule can often be a superior civic to Representation in the early through mid-game. It's also pretty cheap to research or trade for. Rep gives you +3 happy in your 5 biggest cities and +3 beakers per specialist and is medium upkeep. HR gives you +1 happy per military unit stationed in a city, is low upkeep and is the favourite civic of more than a couple of AIs. With HR it is possible to have a size 20 bureauocratic capital in the 6-700 AD range.

Depending on the in-game situation consider building the Oracle for MC or COL can often be superior to the Mids (it's cheaper and lies on the path to HR, another option for the bonus tech) or the Great Lighhouse. If your land dictates a lot of coastal cities and/or there are opportunities to build a couple of offshore cities, TGLH can often outperform early Rep from Pyramids.

Research order is:
*snip*

You may want to consider teching Mysticism yourself. Monuments are a very cheap way to pop the borders of your 2nd-nth cities. Once you are running caste, they aren't needed but Caste is often better delayed. Once Music is in, simply build culture in new/conquered cities until the first border pop.

Also, you may want to consider teching Aesthetics/lit yourself. The AI tends not to prioritize Aesthetics, so it can be traded around for techs that you have bypassed on the way. Literature is a great, cheap tech. It enables the Heroic and National Epics and the GL. Getting the HE and NE built early will greatly boost your military and GP production. The Great Lib is a bit pricey without marble, but can be well worth it for the 8 scientist GPP it provides.


As always, exact tech paths and builds depend on the land you have to work with and the AI's you are up against. Something that works brilliantly in one game is often sub-par in another.

Also, if you have a capital that is suitable for cottage spamming (ie 8+rivers and grass/floodplains) tech Civil Service before machinery. Bureaucracy will get you to machinery a lot faster than machinery will get you to Bureaucracy.

Bear in mind that the above research tree only applies IF i am playing as Toku, and is subject to change pending conditions. It is only an objective guide that I have used to head the ship of state towards domination, not a rigid set of rules. However, for the sake of improving my game, please let me know an improved order for research, and let me know if I am skipping a critical tech. If you can provide the reasons as well, then I will not only have a fish, but will also be able to fish.

Thanks again for the advice!

As I said before, post up a save if you want specific advice. One of the beauties of Civ4 is that it doesn't lend it's self to "cookie cutter" solutions.

[wodan11]

Research order is:

Some nonessential techs in there. e.g., Horseback Riding, Archery

Slavery allows you to convert surplus food into hammers. In a high food surplus city it will increase production significantly.

Unless the city has 3-4 food resources (which is the "except in some few cases" I mentioned), then this isn't the case.

I'm inclined to agree that SP/caste workshops generally perform better.

Even a couple of mines will generally perform better.

Otherwise, excellent advice as usual.

[smbakeresq]

Late to the thread but:

I play Japan all the time, my wife is Japanese and I travel there all the time. I think the problem you are having is a more general one of having a 2:1 ratio of military to wealth/production cities. I think that is way to high. A 1:1 ratio will be better served.

With their traits and Samurai, you dont need more troops to be effective, your troops will be better than what you encounter anyway. You just need to support your quality advantage. With Epic Speed, your Samurai advantage will last a long time, and they as good or better than anything you will encounter in the midgame, so that is the time to expand.

I have had trouble with Japan when I attack too early. Now, barring something irresistable, I wait until the Samurai come along and then triple my holdings rapidly. Until then, I just have 2-3 cities building units until unit cost becomes too high, you can get many free units depending on your civics. All units are spears or swords, with an archer type for every city I intend to capture. With Construction you get your catapults, but your need less of them since the Samurai are so good. I always try to run a little cash surplus and go with all cottages, you will need the money to upgrade.

On Epic, Huge, I find I can have about 20 Samurai, 5-8 spears and some cats to go pillaging with, with more in the queue. Instead of a healer, I just use a Samurai promoted to Woodsman III, that gives them the defense bonus in the woods, healing ability, and 4-7 first strikes. If you stick to the trees when possible it takes a lot of units to were that unit down. I like to use 3 armies, 6-7 Samurai, 2-3 spears, 4-5 cats each, you can pressure three cities that way. If you see a big stack, hit them with cats and then the Samurai even if you are at a numerical disadvantage, you will destroy more than you lose and can just grind them down that way.

The other nations need more units to beat you, since your troops are better. You will capture many good cities this way, let them support you economy. After the unrest of a captured city goes away, I slave defenders and cats to keep going.

I am behind early in techs, prune a few civs with the samurai, then catch up rapidly since I generally have more cities than anyone else. I dont stop with the samurai until there effectiveness passes. By then, you are about even in tech but hold more space and shale plants come online, you can then catch up with the science buildings.

Works for me.